I was reading in some Solidworks 2004 blurb that it now had the ability to
deform shapes by pulling on control points, can anyone point me to the menu
item for this please??
BTW I have looked in the help file under 'deform' maybe this is not the
From what I've read from others who should know, the Deform function
isn't worth fooling with. It's apparently just that bad, and like
PhotoWorks2 never should have been released. Of course, your mileage
Mark 'Sporky' Stapleton
Phil Evans wrote:
Prepare to be disappointed. It (most likely) does not work the way that you
want it to work.
I would be very interested to hear a success story from anybody that has
used the deform feature and got exactly what they needed. I've asked around
and have not heard any success stories, except maybe making some props for
renderings that will never bear the scrutiny of manufacture. This is nit
meant to be a slam - I'm just curious what application it actually has.
Like everything else, it has potential. (add subscription $ here)
Getting good results from Deform is painful, imho.
It becomes a whole other task in how you setup up your curves (edge,
intersection and face curves or points) and hold curves as well as the
different shape options (tweaks).
This is SW Corps (marketing) attempt at GSM (Globe Shape Modeling) but
it does not address the real what/need,.. to directly interact with the
surface topology (with history). It needs to be more fluid.
I freak out on this new stuff because looking at the past and current
issues with SW surface inconsistencies... the chances are my models,
which use deform, will fail in a later sp or release. (beta testing
It is not a NURB type of functionality. It is trying to be an ID tool.
Depending on what you are trying to achieve you can get a much better
deform function using suface fill or do a loft and add a section
post-loft. If you edit the loft section when you first create it, you
get a close proxmity of real time surface deformation. Hopes this
SW is definitely not on the right track with free-form modeling. They
can't be serious about organic modeling until they fix fundamental
issues with curvature continuity control, especially w.r.t. splines.
I don't think SW wants to expose its users to the "innards" of surface
modeling. They want neatly packaged, (relatively) easy to use
features so as not to confuse underschooled users. They have not been
successful at making these features useful.
I haven't used it or seen it used on real product data yet. I've fooled
with it and made some examples, but I'm having difficulty seeing a
practical application. In fact, I'm having some difficulty imagining
where I would want to use the concept, assuming that the tool would do
what you imagined it should do.
It seems another step down the dome-shape-deform path that gets
decreasingly useful and increasingly complex. I do use dome and shape
every time I make a pimple-shaped feature. I know some users who
affectionately refer to the "shape" tool as the "tit wizard". Makes
more sense if you look at the toolbar icon, I guess, and the shapes that
you inevitably wind up with when just messing with it.
The dome-shape-deform philosophy seems to take for granted that SW users
can't make complex shapes and need to have some interface to get in your
way just to prove it. That's one of the things I really like about
ShapeWorks - it really doesn't have much of an interface, and it allows
you to directly maniplulate the surfaces, although it's not very
parametric or associative.
The deform tool also seems to assume that you don't model what you need,
you model something else and then hope this random shape generator will
make it look like you can model complex shapes. That's a little unfair,
I guess. I just wanted to use the phrase "random shape generator" in a
sentence. I have no idea why that term comes to mind when talking about
the deform tool...
It does allow you to take an edge and deform it to another edge or
curve. It won't go to a projected curve, but I found that if the
projected curve is converted into a 3D sketch, it will use that for the
target curve. I was able to get somewhat controlled shapes out of the
curve to curve deform, and only pimple-breast-phallus type shapes out of
the point deform.
It's at least conceivable that you could do something useful with it.
Maybe we should sponsor a contest for the first person to get a design
into production using the deform feature. I'd be interested to see some
practical applications other than sales demos where think3 is the
Hmm. Could we trade in the deform, mold tools and 2D emulator for a
ruled surface that really works?
"Edward T Eaton" wrote in
I haven't actually tried this yet, but I was thinking that it might be
useful for modeling parts that are deformed in their assembled position,
like snaps and sheet metal springs.
Tripod Data Systems
There you go! I think those are excellent suggestions. Basically,
anything that flexes in real life can probably benefit from that
Hmmm, maybe I could use it to animate a mud-flap with one
of those chrome chicks on em, flapping around... or maybe not.
Another example would be o-rings.
In mold design we often use o-rings as static seals between mold plates. In
assembly they are `squeezed'. It amuses me no end that some of the folks
supplying libraries for mold design offer o-rings for our use, but with no
allowance for showing them in the `used state'.
I haven't tried this yet, but perhaps a configuration of the o-rings in the
compressed state could be accomplished with the deform tool.